Probability Distribution
Probability Distribution
Homayora Yeasmin
Lecturer & Acting Head
Department of Agricultural Statistics & Bioinformatics
Khulna Agricultural University, Khulna
Random variable
The variable each of which is associated with a defined probability, it is called random variable.
For example, if 5 coins are tossed, 0,1,3,4 and 5 flowers may appear. The number flower is a
random variable.
Probability distribution
If the values of a random variable are arranged according to their probability, it is called the
probability distribution of the random variable. Probability distribution dictates what we should
expect to observe in a frequency distribution
Let X denoted represents the probabilities of a random variable X by mathematical formula.
Such a formula would necessarily be a function of the numerical values x, that is denoted by f(x),
g(x), ln(x) and so on. Hence we write f(X=x)=P(X=x). If it is a discrete variable then its
probability distribution is called discrete probability distribution and its probability function is
denoted by P(X). On the other hand, if the variable continuous the distribution is called
continuous probability distribution and its probability function is usually expressed by f(X).
According to the law of probability the sum of the probabilities of all values of a variable is 1.
𝑛
i.e., ∑X P(X) = 1 similarly, ∫𝑋 𝑓(𝑋)𝑑𝑋 = 1
Binomial distribution
Let, P be the probability of occurrence and q be the probability of not occurrence of a particular
event in a single trial so that p+q=1. If the experiment is repeated for n independent trials, the
probability of occurrence of an event x times (and not occurring n-x times) may be expressed as
The event of occurrence is sometimes termed as success and not occurrence is termed as failure
and the distribution is known as Binomial distribution.
Swedish Mathematician Bernoulli first discovered the Binomial distribution 1700. The
distribution is also known Bernoulli distribution.
Proof:
Variance = npq
= np(1-p)
=np-n𝑝2
=mean- n𝑝2
(Proved)
Poisson distribution
Let m be the mean or expected number of success in a specifiedtime or spaceand the random
variable X designate the number of success in a given time interval or specified region. Then the
probability distribution of X a Poisson distribution is given by the formula-
𝑒 −𝑚 𝑚 𝑥
𝑃(𝑋) = ; 𝑥 = 0,1,2,3, … , ∞
𝑥!
Where, e is a constant and e=2.71828, m is the only parameter of the distribution and x is any
positive value X can assume.
Examples:
Normal distribution
A random variable x is said to have normal distribution with parameters μ (mean) and σ(standard
deviation) if its density function is given by
1
1 − (𝑥−𝜇)2
𝑓(𝑥) = 𝑒 2𝜎2 ; -∞≤x≤∞
б√2𝜋
Normal distribution has wide application in statistical quality control in industry for
setting control limits.
The entire theory of small sample tests, viz., F, 𝜒 2 , t test etc. is based on the fundamental
assumptions that the percent populations from which the samples have been drawn follow
normal distribution.
Many of the sampling distributions, e.g., student’s t, Snedecor’s F, Ch-square distribution
etc., tends to normality for large samples.